Unexplained Early Neurological Deterioration After Intravenous Thrombolysis

After intravenous recombinant tissue-type plasminogen activator (IV-rtPA) for acute ischemic stroke (AIS), clinical evolution in the first 24 hours is largely unpredictable, underlying the need to better investigate this time window. Although the majority of patients with AIS substantially improve after IV-rtPA, a fraction experience early neurological deterioration (END). Because END consistently predicts poor outcome, it is important to prevent or treat this detrimental event. However, many uncertainties still preclude its informed management. First, estimates of its incidence vary widely, depending on clinical definition and time window used. Second, although some ENDs have a clear cause such as symptomatic intracerebral hemorrhage (sICH), malignant edema, and early recurrent ischemic stroke, leading to evidence-based management, no clear cause is found in the rest, a clinical category sometimes referred to as progressive stroke. This study focuses on this category of post-thrombolysis END for which no underlying mechanism for the worsening of the initial neurological deficit is identified, to be operationally referred to as END unexplained in what follows. There are currently no management guidelines for postthrombolysis unexplained END, and in routine clinical practice, Background and Purpose—Early neurological deterioration (END) after anterior circulation stroke is a serious clinical event strongly associated with poor outcome. Regarding specifically END occurring within 24 hours of intravenous recombinant tissue-type plasminogen activator, apart from definite causes such as symptomatic intracranial hemorrhage and malignant edema whose incidence, predictors, and clinical management are well established, little is known about END without clear mechanism (END unexplained ). Methods—We analyzed 309 consecutive patients thrombolysed intravenously ≤4.5 hours from onset of anterior circulation stroke. END unexplained was defined as a ≥4-point deterioration on 24-hour National Institutes of Health Stroke Scale, without definite mechanism on concomitant imaging. END unexplained and no-END patients were compared for pretreatment clinical and imaging (including magnetic resonance diffusion and diffusion/perfusion mismatch volumes) data and 24hour post-treatment clinical (including blood pressure and glycemic changes) and imaging (24-hour recanalization) data, using univariate logistic regression. Exploratory multivariate analysis was also performed after variable reduction, with bootstrap analysis for internal validation. Results—Among 33 END patients, 23 (7% of whole sample) had END unexplained . END unexplained was associated with poor 3-month outcome (P<0.01). In univariate analysis, admission predictors of END unexplained included no prior use of antiplatelets (P=0.02), lower National Institutes of Health Stroke Scale score (P<0.01), higher glycemia (P=0.03), larger mismatch volume (P=0.03), and proximal occlusion (P=0.01), with consistent results from the multivariate analysis. Among factors recorded during the first 24 hours, only no recanalization was associated with END unexplained in multivariate analysis (P=0.02). Conclusions—END unexplained affected 7% of patients and accounted for most cases of END. Several predictors and associated factors were identified, with important implications regarding underlying mechanisms and potential prevention of this ominous event. (Stroke. 2014;45:2004-2009.)